[mapserver-dev] The legendary story of mapObj.zoomRectangle (a cabaret show)

Alan Boudreault aboudreault at mapgears.com
Fri Mar 5 10:17:22 EST 2010


Fixed in trunk and backported in 5.6:

http://trac.osgeo.org/mapserver/ticket/3286

Alan

On March 4, 2010 12:00:06 pm Alan Boudreault wrote:
> I'm going to revert the mapObj.zoomRectangle change later today or early
> tomorrow.
> 
> On March 4, 2010 11:22:57 am Lime, Steve D (DNR) wrote:
> > I'd like to take a peek at the isValidItem() issue before doing a
> > 5.6.2...
> >
> > -----Original Message-----
> > From: mapserver-dev-bounces at lists.osgeo.org
> >  [mailto:mapserver-dev-bounces at lists.osgeo.org] On Behalf Of Alan
> >  Boudreault Sent: Wednesday, March 03, 2010 3:47 PM
> > To: mapserver-dev at lists.osgeo.org
> > Subject: Re: [mapserver-dev] The legendary story of mapObj.zoomRectangle
> > (a cabaret show)
> >
> > Tamas,
> >
> > Someone also reported on IRC that problem (IIRC, 2-3 weeks ago). My
> > change was a misinterpretation and shouldn't have been done. I had not
> > understood why this part of the code was different of the php similar
> > function. As you said on IRC "treating maxy >= miny as a failure is
> > semantically irregular", but this commit changed the behavior of the
> > function that exists for a long time. I would suggest to revert it and
> > backport it in branch 5.6.
> >
> > Perhaps it would be a good opportunity to release 5.6.2 ? Are there
> > enough important fixes in the branch?
> >
> > regards,
> > Alan
> >
> > On March 3, 2010 03:49:54 pm Tamas Szekeres wrote:
> > > Folks,
> > >
> > > Upon upgrading my existing application to 5.6.1 I'm getting frustrated
> > >  about the changes of the rectangle check in mapObj.zoomRectangle. So
> > > let me summarize how the related code have changed within the past 5
> > > years.
> > >
> > > - 2004-06-26 The stuff has been committed by sean in r3294 mapzoom.i Ln
> > > 280 if (poPixRect->maxy >= poPixRect->miny)
> > >         {
> > >             msSetError(MS_MISCERR, "Georeferenced coordinates miny <=
> > >  maxy", "mapscript::mapObj::zoomRectangle()");
> > >             return MS_FAILURE;
> > >         }
> > >
> > > - 2006-05-30  First attempt to swap miny and maxy by umberto in r5472
> > >
> > > - 2006-07-24  The change have been reverted by hobu in r5542
> > >
> > > - 2006-12-13 Second attempt to swap miny and maxy by sdlime in r5859
> > >
> > > - 2006-12-13 Reverted the change by sdlime in r5864
> > >
> > > - 2009-10-12 Third attempt to swap miny and maxy by aboudreault in
> > > r9444
> > >
> > > There have been a couple of tickets related to this issue like:
> > > http://trac.osgeo.org/mapserver/ticket/1817
> > > http://trac.osgeo.org/mapserver/ticket/1982
> > > http://trac.osgeo.org/mapserver/ticket/1988
> > > http://trac.osgeo.org/mapserver/ticket/3286
> > >
> > >
> > > While I'm a bit lost with the reason of these reverts it seems we don't
> > >  have solid standpoint on how this consistency check should be done. I
> > >  would anticipate I'm not against this change, however I would not want
> > > to alter my existing scipts (like now) in every release in order to
> > > have my code working.
> > >
> > > Could someone confirm this change won't be reverted in the future
> > > releases?
> > >
> > >
> > > Best regards,
> > >
> > > Tamas
> 

-- 
Alan Boudreault
Mapgears
http://www.mapgears.com


More information about the mapserver-dev mailing list